3 minute read
Figure 2-8: High-Injury Network Method
2.5 Collision Mapping Overview
Understanding the roadway conditions where severe-injury and fatal collisions occur is fundamental to addressing unsafe conditions in other places before these types of collisions occur. Mapping what is known as the “high-injury network” identifies segments of roadway that have been the location of the highest number of pedestrian-involved, bicyclist-involved, severe-injury or fatal collisions. By learning the characteristics of high-injury road segments, these conditions can be remedied citywide where clusters of collisions occur. A “collision tree” grouping method helps identify clusters of collisions that occur in the same conditions as the high-injury network.
Advertisement
Collision mapping was performed in the following two ways:
1. A high-injury network method that ranks road segments by weighting the collisions that occurred on each segment over the study period 2. A collision tree method that groups and maps collisions by the roadway conditions where they occurred.
2.5.1 High-Injury Network
The high-injury network (HIN) method identifies segments of roads with concentrations of fatal and severe injury (Killed or Significantly Injured-KSI) collisions. The goal of this method is to identify connected segments of roadway with a high number of collisions that together form a high-injury corridor where safety improvements can be implemented. With this method, bicyclist-involved and pedestrian-involved collisions are weighted 25% more than auto-only collisions. Fatal auto-only collisions are also weighted 25% more than non-fatal collisions. Figure 2-8 shows the process for identifying high-injury segi ments. For both methods, larger sums indicate more collisions, or a sum of relatively fewer auto-only collisions combined with bicycle- or pedestrian-involved collisions on the same road segment.
Figure 2-8: High-Injury Network Method
Figure 2-9 shows the mapped result of the method shown in Figure 2-8. The map legend shows that a HIN “score” (or KSI, the sum of fatal and severe collisions per road segment) can be translated to a collision type or combination of collision types. The HIN score is normalized (divided) by the length of the respective road segment and expressed as a percentile compared to all road segments in El Monte (Figure 2-10). Road segments from the 75th to 100th percentiles have the highest number of pedestrian-involved, bicyclist-involved, severe-injury or fatal collisions for their length compared to all road segments citywide. Figure 2-10 shows disparate segments make up the high-ins jury network, from which it is difficult to identify corridors for safety improvements. Examples of these disconnected high-injury segments exist in northern El Monte at Peck Road and Lower Azusa Road, and in southern El Monte at Peck Road and Garvey Avenue.
To help connect the segments, the same method was applied to a larger collision dataset that includes minor injury collisions. Figure 2-11 shows the results of this larger dataset, summarized as the “collision index” (CI) to differentiate it from the high-injury network. Figure 2-12 shows the results of the collision index normalized by segment length, as was done for the high-injury network in Figure 2-10. This method identifies more road segments that together link the segments from the high-injury network to identify the broader corridors for safety improvements. The project Technical Advisory Committee (TAC; described further in Chapter 3) identified corridors of concern that correlate to the corridors identified by CI method. The CI method identifies parts of most of the TAC corridors of concern, and helps identify corridors that the TAC did not identify, such as Peck Road and Ramona Boulevard. The TAC also identified Parkway Drive (south of Durfee Avenue in southern El Monte) and segments of Durfee Avenue north of Ramona Boulevard. These segments were likely identified by the TAC because of concerns with vehicle speeds in residential neighborhoods, but collisions were either low in number or non-existent. Though the HIN method did not include these segments, the combination of this data-driven CI method together with community input from the TAC help give a comprehensive picture of the corridors that can be focused on for safety improvements.
